Rag And Retrieval easy
Users of a policy chatbot ask follow-ups such as 'what about for part-time staff?' Retrieval on that text alone returns irrelevant chunks. What helps most?
- A Rewrite the follow-up into a standalone query using chat history before retrieving.
- B Embed the entire conversation transcript as one query vector on every turn.
- C Raise top-k so that some relevant chunks still appear despite the vague query.
- D Skip retrieval on follow-up turns and reuse the chunks fetched for the first question.
Answer: A. Rewrite the follow-up into a standalone query using chat history before retrieving.
**Query rewriting** turns 'what about part-time staff?' into 'parental leave policy for part-time staff', giving retrieval a precise target at the cost of one extra fast LLM call. Option B is tempting, but embedding a long transcript blurs the current intent across many earlier topics. Option D works only when the follow-up truly stays within the same documents.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You must index 100M chunks using a 3,072-dimension embedding in fp32, served from RAM via HNSW. The raw vectors alone take about 1.2 TB before graph overhead. You need to cut memory several-fold while losing at most a point or two of recall@10. Which approach fits?
- A multi-tenant SaaS app stores all tenants in one vector index with a tenant_id field. Queries fetch the top 20 chunks by similarity and then drop chunks from other tenants. Small tenants often get 0-2 results. Why does this happen, and what is the fix?
- Runbooks are split into fixed 300-token chunks with no overlap. When a procedure step starts near the end of one chunk and finishes in the next, answers come back incomplete. What is the lowest-cost fix?
- Your offline eval shows recall@5 of 92% on a labeled set, so the gold passage is in the context for nearly every question. Yet human graders mark 35% of answers wrong, and the wrong answers often ignore or contradict the gold passage. Where should the team invest next?
- Your product requires every RAG answer to include clickable source citations that users can verify. How should those citations be produced?
- A runbook RAG system uses fixed 500-token chunks with no overlap. Users who ask for a multi-step recovery procedure get answers that stop at step 4 of 9, and retrieval logs show the top hit ends mid-procedure. What is the most targeted fix?